How Alaska's Senate Race Became a Toss-Up
A state Trump carried by fourteen points is now rated a coin flip. The topline poll numbers alone don't explain why — the crosstabs, the approval numbers, and a ballot fight over two men named Dan Sullivan do.
On paper, this shouldn't be close. Alaska backed Donald Trump by roughly fourteen points in 2024, and Republican Sen. Dan Sullivan won re-election in 2020 by nearly thirteen. Yet the Cook Political Report and Sabato's Crystal Ball both now list the seat as a toss-up, and the most recent New York Times/Siena College poll puts Sullivan ahead of Democratic challenger Mary Peltola by just two points — 47 to 45 percent.
That two-point margin is the number making headlines. It is also, on its own, a poor guide to what's actually happening beneath it.
01 · The Independent Paradox
The same NYT/Siena poll that shows a two-point race also shows Peltola dominating among voters who identify as independent — a bloc that, in the poll's Party ID crosstab, makes up 44 percent of the electorate, larger than either party's self-identifiers.
Weight that crosstab against the poll's own party breakdown — 14 percent Democrat, 29 percent Republican, 44 percent independent — and it implies a Peltola topline closer to the high 30s, not 45. The gap between that math and the reported number is real, and it comes down to a distinction worth making explicit: party ID is not party registration. Registration is a fixed administrative fact; in Alaska, a majority of voters are registered nonpartisan or undeclared by default. Party ID is a live, self-reported answer to "how do you think of yourself politically right now." Pollsters build likely-voter models differently depending on which one they lean on, and that choice — not any single subgroup number — is usually where a topline and a crosstab quietly diverge.
The gap doesn't mean the poll is wrong. It means the topline and the independent crosstab are answering two different questions.
02 · Trump's Alaska Problem
One reason the independent number matters so much: those voters are not neutral on the president, and Alaska is not the state it was in November 2024.
| Party ID | Net Approve | Net Disapprove |
|---|---|---|
| Republican | 90% | 9% |
| Independent | 36% | 60% |
| Democrat | 13% | 87% |
Republicans remain solidly behind the president; the erosion is concentrated almost entirely in the independent lane. Two separate national trackers have found Trump's net approval in Alaska specifically underwater — Civiqs at -8, Morning Consult at -5, putting the state in similar territory to Ohio and behind only North Carolina, Iowa, and Georgia among competitive 2026 states. For a state Trump won by fourteen points eighteen months ago, that is a meaningful shift, and it lands directly on an incumbent who has voted with national party leadership on the large majority of roll-call votes.
03 · A Turnout Wildcard, Not a Turnout Wave
If independents are souring on the president and breaking for Peltola by double digits, the obvious follow-up question is whether they'll actually show up. The polling suggests caution on that point.
| Group | High-propensity |
|---|---|
| Democrat | 91% |
| Republican | 83% |
| Other / Independent | 71% |
By this measure, independents report lower vote-certainty than either party's base — not higher. Regionally, the pattern has some texture worth noting: Alaska's most reliably conservative areas, the Kenai Peninsula and Mat-Su Valley, show softer "almost certain" numbers (49%) than the more centrist-to-Democratic Anchorage (60%) and Southeast (66%). That's consistent with — though it doesn't prove — a story where some conservative-leaning voters are disengaged rather than newly persuaded. The lowest overall certainty is in North/West Alaska (44%), a largely rural, Alaska Native region where lower measured propensity is a long-running pattern independent of ideology.
04 · The "Yes Man" Argument
Into that opening has stepped an outside messaging effort. The 907 Initiative, an Alaska-based watchdog group, has been running a campaign branded "Yes Man Dan," arguing Sullivan votes with national Republican leadership roughly 98 percent of the time — on tariffs, Medicaid, and SNAP among other issues — even when it cuts against Alaska-specific interests. It is not Peltola's campaign making this case directly; it's a third-party frame designed to work on exactly the voters shown souring on Trump without yet moving to Peltola: the idea that Sullivan answers to Washington first.
05 · Two Dan Sullivans
That argument got an unplanned assist in June, when Dan J. Sullivan, a 69-year-old retired teacher from Petersburg, filed to run for the same seat as the incumbent — as a Republican.
Division of Elections Director Carol Beecher disqualified him on June 15, finding his candidacy was not filed in "good faith." Sen. Sullivan's campaign called him a "sham candidate" and alleged Democratic coordination; the National Republican Senatorial Committee and the Alaska Republican Party filed complaints, and attorneys general from fourteen Republican-led states filed briefs supporting his removal. The state hired an outside law firm — Colorado-based First and Fourteenth, described in reporting as a firm that "takes on conservative and Republican causes" — for up to $100,000 to argue the case. Peltola's campaign and Alaska Democrats denied any involvement; Beecher's own disqualification order cited no evidence of coordination, only circumstantial ties.
The Alaska Supreme Court sided with the challenger, unanimously among participating justices, and remanded the question of ballot listing back to the Division "within the confines of existing Alaska ballot design law." The sample ballot that followed listed the challenger as "Sullivan, Daniel J. Jr." with no party affiliation shown — despite his being a registered Republican, and despite every other candidate on the ballot having a party listed — while the incumbent was labeled "Sullivan, Dan S. (Registered Republican) Incumbent," an incumbency label with no clear precedent on an Alaska primary ballot. Independent legal commentary flagged the listing as inconsistent with the law on three separate points.
Whatever the merits of the underlying "sham candidacy" dispute — and reasonable people involved in the case disagree sharply on that question — the optics of the sequence are straightforward: an incumbent senator, his party, the NRSC, and fourteen state attorneys general spent weeks fighting to keep a fellow Republican off the ballot, lost at the state's highest court, and then produced a follow-up document that appeared not to fully comply with that court's instructions. Peltola was not a party to any of it.
06 · Where the Other Polls Land
The NYT/Siena topline (Sullivan +2) is not the only recent read on the race. Alaska Survey Research, an in-state pollster, has shown Peltola ahead in its last two surveys:
| Field dates | Peltola | Sullivan |
|---|---|---|
| April 2026 | 49.0% | 42.5% |
| Jun 4–7, 2026 | 49.4% | 44.2% |
Different pollsters use different likely-voter models; the spread between surveys is itself informative about how unsettled the race is, not evidence that any single poll is wrong.
07 · What to Watch
- Whether the ballot listing dispute continues. The challenger's attorney has not ruled out further legal action over how the Division ultimately printed the ballot.
- Post-ruling polling. The NYT/Siena field dates predate the Supreme Court's ruling and the sample-ballot controversy; whether independents' views of Sullivan move at all will only show up in polling taken after June 29–30.
- Primary results, August 18. Prediction markets currently give the incumbent Sullivan a 93 percent chance of advancing through the top-four nonpartisan primary, and Peltola 94 percent — the more interesting question is likely the margins each carries into a ranked-choice general.
- Whether "Yes Man Dan" messaging gains traction among the specific slice of conservative-leaning independents who disapprove of Trump but haven't yet said they'd vote for Peltola.
Posts
No comments:
Post a Comment